Predictive models of miscarriage on the basis of data from a preconception cohort study.

OBJECTIVE: To use self-reported preconception data to derive models that predict the risk of miscarriage. DESIGN: Prospective preconception cohort study. SETTING: Not applicable. PATIENTS: Study participants were female, aged 21-45 years, residents of the United States or Canada, and attempting spontaneous pregnancy at enrollment during 2013-2022. Participants were followed for up to 12 months of pregnancy attempts; those who conceived were followed through pregnancy and postpartum. We restricted analyses to participants who conceived during the study period. EXPOSURE: On baseline and follow-up questionnaires completed every 8 weeks until pregnancy, we collected self-reported data on sociodemographic factors, reproductive history, lifestyle, anthropometrics, diet, medical history, and male partner characteristics. We included 160 potential predictor variables in our models. MAIN OUTCOME MEASURES: The primary outcome was a miscarriage, defined as pregnancy loss before 20 weeks of gestation. We followed participants from their first positive pregnancy test until miscarriage or a censoring event (induced abortion, ectopic pregnancy, loss of follow-up, or 20 weeks of gestation), whichever occurred first. We fit both survival and static models using Cox proportional hazards models, logistic regression, support vector machines, gradient-boosted trees, and random forest algorithms. We evaluated model performance using the concordance index (survival models) and the weighted F1 score (static models). RESULTS: Among the 8,720 participants who conceived, 20.4% reported miscarriage. In multivariable models, the strongest predictors of miscarriage were female age, history of miscarriage, and male partner age. The weighted F1 score ranged from 73%-89% for static models and the concordance index ranged from 53%-56% for survival models, indicating better discrimination for the static models compared with the survival models (i.e., the ability of the model to discriminate between individuals with and without miscarriage). No appreciable differences were observed across strata of miscarriage history or among models restricted to ≥8 weeks of gestation. CONCLUSION: Our findings suggest that miscarriage is not easily predicted on the basis of preconception lifestyle characteristics and that advancing age and a history of miscarriage are the most important predictors of incident miscarriage.

Multi-Attribute Subset Selection enables prediction of representative phenotypes across microbial populations.

The interpretation of complex biological datasets requires the identification of representative variables that describe the data without critical information loss. This is particularly important in the analysis of large phenotypic datasets (phenomics). Here we introduce Multi-Attribute Subset Selection (MASS), an algorithm which separates a matrix of phenotypes (e.g., yield across microbial species and environmental conditions) into predictor and response sets of conditions. Using mixed integer linear programming, MASS expresses the response conditions as a linear combination of the predictor conditions, while simultaneously searching for the optimally descriptive set of predictors. We apply the algorithm to three microbial datasets and identify environmental conditions that predict phenotypes under other conditions, providing biologically interpretable axes for strain discrimination. MASS could be used to reduce the number of experiments needed to identify species or to map their metabolic capabilities. The generality of the algorithm allows addressing subset selection problems in areas beyond biology.

ITNR: Inversion Transformer-based Neural Ranking for cancer drug recommendations.

Personalized drug response prediction is an approach for tailoring effective therapeutic strategies for patients based on their tumors' genomic characterization. While machine learning methods are widely employed in the literature, they often struggle to capture drug-cell line relations across various cell lines. In addressing this challenge, our study introduces a novel listwise Learning-to-Rank (LTR) model named Inversion Transformer-based Neural Ranking (ITNR). ITNR utilizes genomic features and a transformer architecture to decipher functional relationships and construct models that can predict patient-specific drug responses. Our experiments were conducted on three major drug response data sets, showing that ITNR reliably and consistently outperforms state-of-the-art LTR models.

Predicting polycystic ovary syndrome with machine learning algorithms from electronic health records.

Introduction: Predictive models have been used to aid early diagnosis of PCOS, though existing models are based on small sample sizes and limited to fertility clinic populations. We built a predictive model using machine learning algorithms based on an outpatient population at risk for PCOS to predict risk and facilitate earlier diagnosis, particularly among those who meet diagnostic criteria but have not received a diagnosis. Methods: This is a retrospective cohort study from a SafetyNet hospital's electronic health records (EHR) from 2003-2016. The study population included 30,601 women aged 18-45 years without concurrent endocrinopathy who had any visit to Boston Medical Center for primary care, obstetrics and gynecology, endocrinology, family medicine, or general internal medicine. Four prediction outcomes were assessed for PCOS. The first outcome was PCOS ICD-9 diagnosis with additional model outcomes of algorithm-defined PCOS. The latter was based on Rotterdam criteria and merging laboratory values, radiographic imaging, and ICD data from the EHR to define irregular menstruation, hyperandrogenism, and polycystic ovarian morphology on ultrasound. Results: We developed predictive models using four machine learning methods: logistic regression, supported vector machine, gradient boosted trees, and random forests. Hormone values (follicle-stimulating hormone, luteinizing hormone, estradiol, and sex hormone binding globulin) were combined to create a multilayer perceptron score using a neural network classifier. Prediction of PCOS prior to clinical diagnosis in an out-of-sample test set of patients achieved an average AUC of 85%, 81%, 80%, and 82%, respectively in Models I, II, III and IV. Significant positive predictors of PCOS diagnosis across models included hormone levels and obesity; negative predictors included gravidity and positive bHCG. Conclusion: Machine learning algorithms were used to predict PCOS based on a large at-risk population. This approach may guide early detection of PCOS within EHR-interfaced populations to facilitate counseling and interventions that may reduce long-term health consequences. Our model illustrates the potential benefits of an artificial intelligence-enabled provider assistance tool that can be integrated into the EHR to reduce delays in diagnosis. However, model validation in other hospital-based populations is necessary.

Fusion of Low-Level Descriptors of Digital Voice Recordings for Dementia Assessment.

Digital voice recordings can offer affordable, accessible ways to evaluate behavior and function. We assessed how combining different low-level voice descriptors can evaluate cognitive status. Using voice recordings from neuropsychological exams at the Framingham Heart Study, we developed a machine learning framework fusing spectral, prosodic, and sound quality measures early in the training cycle. The model's area under the receiver operating characteristic curve was 0.832 (±0.034) in differentiating persons with dementia from those who had normal cognition. This offers a data-driven framework for analyzing minimally processed voice recordings for cognitive assessment, highlighting the value of digital technologies in disease detection and intervention.

Improved prediction of MHC-peptide binding using protein language models.

Major histocompatibility complex Class I (MHC-I) molecules bind to peptides derived from intracellular antigens and present them on the surface of cells, allowing the immune system (T cells) to detect them. Elucidating the process of this presentation is essential for regulation and potential manipulation of the cellular immune system. Predicting whether a given peptide binds to an MHC molecule is an important step in the above process and has motivated the introduction of many computational approaches to address this problem. NetMHCPan, a pan-specific model for predicting binding of peptides to any MHC molecule, is one of the most widely used methods which focuses on solving this binary classification problem using shallow neural networks. The recent successful results of Deep Learning (DL) methods, especially Natural Language Processing (NLP-based) pretrained models in various applications, including protein structure determination, motivated us to explore their use in this problem. Specifically, we consider the application of deep learning models pretrained on large datasets of protein sequences to predict MHC Class I-peptide binding. Using the standard performance metrics in this area, and the same training and test sets, we show that our models outperform NetMHCpan4.1, currently considered as the-state-of-the-art.

Predicting polycystic ovary syndrome (PCOS) with machine learning algorithms from electronic health records.

Introduction: Predictive models have been used to aid early diagnosis of PCOS, though existing models are based on small sample sizes and limited to fertility clinic populations. We built a predictive model using machine learning algorithms based on an outpatient population at risk for PCOS to predict risk and facilitate earlier diagnosis, particularly among those who meet diagnostic criteria but have not received a diagnosis. Methods: This is a retrospective cohort study from a SafetyNet hospital's electronic health records (EHR) from 2003-2016. The study population included 30,601 women aged 18-45 years without concurrent endocrinopathy who had any visit to Boston Medical Center for primary care, obstetrics and gynecology, endocrinology, family medicine, or general internal medicine. Four prediction outcomes were assessed for PCOS. The first outcome was PCOS ICD-9 diagnosis with additional model outcomes of algorithm-defined PCOS. The latter was based on Rotterdam criteria and merging laboratory values, radiographic imaging, and ICD data from the EHR to define irregular menstruation, hyperandrogenism, and polycystic ovarian morphology on ultrasound. Results: We developed predictive models using four machine learning methods: logistic regression, supported vector machine, gradient boosted trees, and random forests. Hormone values (follicle-stimulating hormone, luteinizing hormone, estradiol, and sex hormone binding globulin) were combined to create a multilayer perceptron score using a neural network classifier. Prediction of PCOS prior to clinical diagnosis in an out-of-sample test set of patients achieved AUC of 85%, 81%, 80%, and 82%, respectively in Models I, II, III and IV. Significant positive predictors of PCOS diagnosis across models included hormone levels and obesity; negative predictors included gravidity and positive bHCG. Conclusions: Machine learning algorithms were used to predict PCOS based on a large at-risk population. This approach may guide early detection of PCOS within EHR-interfaced populations to facilitate counseling and interventions that may reduce long-term health consequences. Our model illustrates the potential benefits of an artificial intelligence-enabled provider assistance tool that can be integrated into the EHR to reduce delays in diagnosis. However, model validation in other hospital-based populations is necessary.

Personalized hypertension treatment recommendations by a data-driven model.

BACKGROUND: Hypertension is a prevalent cardiovascular disease with severe longer-term implications. Conventional management based on clinical guidelines does not facilitate personalized treatment that accounts for a richer set of patient characteristics. METHODS: Records from 1/1/2012 to 1/1/2020 at the Boston Medical Center were used, selecting patients with either a hypertension diagnosis or meeting diagnostic criteria (≥ 130 mmHg systolic or ≥ 90 mmHg diastolic, n = 42,752). Models were developed to recommend a class of antihypertensive medications for each patient based on their characteristics. Regression immunized against outliers was combined with a nearest neighbor approach to associate with each patient an affinity group of other patients. This group was then used to make predictions of future Systolic Blood Pressure (SBP) under each prescription type. For each patient, we leveraged these predictions to select the class of medication that minimized their future predicted SBP. RESULTS: The proposed model, built with a distributionally robust learning procedure, leads to a reduction of 14.28 mmHg in SBP, on average. This reduction is 70.30% larger than the reduction achieved by the standard-of-care and 7.08% better than the corresponding reduction achieved by the 2nd best model which uses ordinary least squares regression. All derived models outperform following the previous prescription or the current ground truth prescription in the record. We randomly sampled and manually reviewed 350 patient records; 87.71% of these model-generated prescription recommendations passed a sanity check by clinicians. CONCLUSION: Our data-driven approach for personalized hypertension treatment yielded significant improvement compared to the standard-of-care. The model implied potential benefits of computationally deprescribing and can support situations with clinical equipoise.

Distributionally robust learning-to-rank under the Wasserstein metric.

Despite their satisfactory performance, most existing listwise Learning-To-Rank (LTR) models do not consider the crucial issue of robustness. A data set can be contaminated in various ways, including human error in labeling or annotation, distributional data shift, and malicious adversaries who wish to degrade the algorithm's performance. It has been shown that Distributionally Robust Optimization (DRO) is resilient against various types of noise and perturbations. To fill this gap, we introduce a new listwise LTR model called Distributionally Robust Multi-output Regression Ranking (DRMRR). Different from existing methods, the scoring function of DRMRR was designed as a multivariate mapping from a feature vector to a vector of deviation scores, which captures local context information and cross-document interactions. In this way, we are able to incorporate the LTR metrics into our model. DRMRR uses a Wasserstein DRO framework to minimize a multi-output loss function under the most adverse distributions in the neighborhood of the empirical data distribution defined by a Wasserstein ball. We present a compact and computationally solvable reformulation of the min-max formulation of DRMRR. Our experiments were conducted on two real-world applications: medical document retrieval and drug response prediction, showing that DRMRR notably outperforms state-of-the-art LTR models. We also conducted an extensive analysis to examine the resilience of DRMRR against various types of noise: Gaussian noise, adversarial perturbations, and label poisoning. Accordingly, DRMRR is not only able to achieve significantly better performance than other baselines, but it can maintain a relatively stable performance as more noise is added to the data.

Social determinants of health and the prediction of missed breast imaging appointments.

BACKGROUND: Predictive models utilizing social determinants of health (SDH), demographic data, and local weather data were trained to predict missed imaging appointments (MIA) among breast imaging patients at the Boston Medical Center (BMC). Patients were characterized by many different variables, including social needs, demographics, imaging utilization, appointment features, and weather conditions on the date of the appointment. METHODS: This HIPAA compliant retrospective cohort study was IRB approved. Informed consent was waived. After data preprocessing steps, the dataset contained 9,970 patients and 36,606 appointments from 1/1/2015 to 12/31/2019. We identified 57 potentially impactful variables used in the initial prediction model and assessed each patient for MIA. We then developed a parsimonious model via recursive feature elimination, which identified the 25 most predictive variables. We utilized linear and non-linear models including support vector machines (SVM), logistic regression (LR), and random forest (RF) to predict MIA and compared their performance. RESULTS: The highest-performing full model is the nonlinear RF, achieving the highest Area Under the ROC Curve (AUC) of 76% and average F1 score of 85%. Models limited to the most predictive variables were able to attain AUC and F1 scores comparable to models with all variables included. The variables most predictive of missed appointments included timing, prior appointment history, referral department of origin, and socioeconomic factors such as household income and access to caregiving services. CONCLUSIONS: Prediction of MIA with the data available is inherently limited by the complex, multifactorial nature of MIA. However, the algorithms presented achieved acceptable performance and demonstrated that socioeconomic factors were useful predictors of MIA. In contrast with non-modifiable demographic factors, we can address SDH to decrease the incidence of MIA.

Automated detection of mild cognitive impairment and dementia from voice recordings: A natural language processing approach.

INTRODUCTION: Automated computational assessment of neuropsychological tests would enable widespread, cost-effective screening for dementia. METHODS: A novel natural language processing approach is developed and validated to identify different stages of dementia based on automated transcription of digital voice recordings of subjects' neuropsychological tests conducted by the Framingham Heart Study (n = 1084). Transcribed sentences from the test were encoded into quantitative data and several models were trained and tested using these data and the participants' demographic characteristics. RESULTS: Average area under the curve (AUC) on the held-out test data reached 92.6%, 88.0%, and 74.4% for differentiating Normal cognition from Dementia, Normal or Mild Cognitive Impairment (MCI) from Dementia, and Normal from MCI, respectively. DISCUSSION: The proposed approach offers a fully automated identification of MCI and dementia based on a recorded neuropsychological test, providing an opportunity to develop a remote screening tool that could be adapted easily to any language.

Predictive models of pregnancy based on data from a preconception cohort study.

STUDY QUESTION: Can we derive adequate models to predict the probability of conception among couples actively trying to conceive? SUMMARY ANSWER: Leveraging data collected from female participants in a North American preconception cohort study, we developed models to predict pregnancy with performance of ∼70% in the area under the receiver operating characteristic curve (AUC). WHAT IS KNOWN ALREADY: Earlier work has focused primarily on identifying individual risk factors for infertility. Several predictive models have been developed in subfertile populations, with relatively low discrimination (AUC: 59-64%). STUDY DESIGN, SIZE, DURATION: Study participants were female, aged 21-45 years, residents of the USA or Canada, not using fertility treatment, and actively trying to conceive at enrollment (2013-2019). Participants completed a baseline questionnaire at enrollment and follow-up questionnaires every 2 months for up to 12 months or until conception. We used data from 4133 participants with no more than one menstrual cycle of pregnancy attempt at study entry. PARTICIPANTS/MATERIALS, SETTING, METHODS: On the baseline questionnaire, participants reported data on sociodemographic factors, lifestyle and behavioral factors, diet quality, medical history and selected male partner characteristics. A total of 163 predictors were considered in this study. We implemented regularized logistic regression, support vector machines, neural networks and gradient boosted decision trees to derive models predicting the probability of pregnancy: (i) within fewer than 12 menstrual cycles of pregnancy attempt time (Model I), and (ii) within 6 menstrual cycles of pregnancy attempt time (Model II). Cox models were used to predict the probability of pregnancy within each menstrual cycle for up to 12 cycles of follow-up (Model III). We assessed model performance using the AUC and the weighted-F1 score for Models I and II, and the concordance index for Model III. MAIN RESULTS AND THE ROLE OF CHANCE: Model I and II AUCs were 70% and 66%, respectively, in parsimonious models, and the concordance index for Model III was 63%. The predictors that were positively associated with pregnancy in all models were: having previously breastfed an infant and using multivitamins or folic acid supplements. The predictors that were inversely associated with pregnancy in all models were: female age, female BMI and history of infertility. Among nulligravid women with no history of infertility, the most important predictors were: female age, female BMI, male BMI, use of a fertility app, attempt time at study entry and perceived stress. LIMITATIONS, REASONS FOR CAUTION: Reliance on self-reported predictor data could have introduced misclassification, which would likely be non-differential with respect to the pregnancy outcome given the prospective design. In addition, we cannot be certain that all relevant predictor variables were considered. Finally, though we validated the models using split-sample replication techniques, we did not conduct an external validation study. WIDER IMPLICATIONS OF THE FINDINGS: Given a wide range of predictor data, machine learning algorithms can be leveraged to analyze epidemiologic data and predict the probability of conception with discrimination that exceeds earlier work. STUDY FUNDING/COMPETING INTEREST(S): The research was partially supported by the U.S. National Science Foundation (under grants DMS-1664644, CNS-1645681 and IIS-1914792) and the National Institutes for Health (under grants R01 GM135930 and UL54 TR004130). In the last 3 years, L.A.W. has received in-kind donations for primary data collection in PRESTO from FertilityFriend.com, Kindara.com, Sandstone Diagnostics and Swiss Precision Diagnostics. L.A.W. also serves as a fibroid consultant to AbbVie, Inc. The other authors declare no competing interests. TRIAL REGISTRATION NUMBER: N/A.

Informative predictors of pregnancy after first IVF cycle using eIVF practice highway electronic health records.

The aim of this study is to determine the most informative pre- and in-cycle variables for predicting success for a first autologous oocyte in-vitro fertilization (IVF) cycle. This is a retrospective study using 22,413 first autologous oocyte IVF cycles from 2001 to 2018. Models were developed to predict pregnancy following an IVF cycle with a fresh embryo transfer. The importance of each variable was determined by its coefficient in a logistic regression model and the prediction accuracy based on different variable sets was reported. The area under the receiver operating characteristic curve (AUC) on a validation patient cohort was the metric for prediction accuracy. Three factors were found to be of importance when predicting IVF success: age in three groups (38-40, 41-42, and above 42 years old), number of transferred embryos, and number of cryopreserved embryos. For predicting first-cycle IVF pregnancy using all available variables, the predictive model achieved an AUC of 68% + /- 0.01%. A parsimonious predictive model utilizing age (38-40, 41-42, and above 42 years old), number of transferred embryos, and number of cryopreserved embryos achieved an AUC of 65% + /- 0.01%. The proposed models accurately predict a single IVF cycle pregnancy outcome and identify important predictive variables associated with the outcome. These models are limited to predicting pregnancy immediately after the IVF cycle and not live birth. These models do not include indicators of multiple gestation and are not intended for clinical application.

Machine Learning for Pharmacogenomics and Personalized Medicine: A Ranking Model for Drug Sensitivity Prediction.

It is infeasible to test many different chemotherapy drugs on actual patients in large clinical trials, which motivates computational methods with the ability to learn and exploit associations between drug effectiveness and patient characteristics. This work proposes a machine learning approach to infer robust predictors of drug responses from patient genomic information. Rather than predicting the exact drug response on a given cell line, we introduce an elastic-net regression methodology to compare a drug-cell line pair against an alternative pair. Using predicted pairwise comparisons we rank the effectiveness of different drugs on the same cell line. A total of 173 cell lines and 100 drug responses were used in various settings for training and testing the proposed models. By comparing our approach against twelve baseline methods, we demonstrate that it outperforms the state-of-the-art methods in the literature. In contrast to most other methods, the algorithm is able to maintain its high performance even when we use a large number of drugs and few cell lines.

Routing and Rebalancing Intermodal Autonomous Mobility-on-Demand Systems in Mixed Traffic.

This paper studies congestion-aware route-planning policies for intermodal Autonomous Mobility-on-Demand (AMoD) systems, whereby a fleet of autonomous vehicles provides on-demand mobility jointly with public transit under mixed traffic conditions (consisting of AMoD and private vehicles). First, we devise a network flow model to jointly optimize the AMoD routing and rebalancing strategies in a congestion-aware fashion by accounting for the endogenous impact of AMoD flows on travel time. Second, we capture the effect of exogenous traffic stemming from private vehicles adapting to the AMoD flows in a user-centric fashion by leveraging a sequential approach. Since our results are in terms of link flows, we then provide algorithms to retrieve the explicit recommended routes to users. Finally, we showcase our framework with two case-studies considering the transportation sub-networks in Eastern Massachusetts and New York City, respectively. Our results suggest that for high levels of demand, pure AMoD travel can be detrimental due to the additional traffic stemming from its rebalancing flows. However, blending AMoD with public transit, walking and micromobility options can significantly improve the overall system performance by leveraging the high-throughput of public transit combined with the flexibility of walking and micromobility.

A Sharp Estimate on the Transient Time of Distributed Stochastic Gradient Descent.

This paper is concerned with minimizing the average of n cost functions over a network in which agents may communicate and exchange information with each other. We consider the setting where only noisy gradient information is available. To solve the problem, we study the distributed stochastic gradient descent (DSGD) method and perform a non-asymptotic convergence analysis. For strongly convex and smooth objective functions, in expectation, DSGD asymptotically achieves the optimal network independent convergence rate compared to centralized stochastic gradient descent (SGD). Our main contribution is to characterize the transient time needed for DSGD to approach the asymptotic convergence rate. Moreover, we construct a "hard" optimization problem that proves the sharpness of the obtained result. Numerical experiments demonstrate the tightness of the theoretical results.

HPC Data Center Participation in Demand Response: an Adaptive Policy with QoS Assurance.

Demand response programs help stabilize the electricity grid by providing monetary stimulus to consumers if they regulate their power consumption following market requirements. Regulation service, a market that requires participants to regulate power by following a signal updated every few seconds, is particularly beneficial to HPC data centers since data centers are capable of increasing/decreasing power consumption owing to the flexibility in running workloads and the availability of power control mechanisms. While prior works have explored how data centers can provide regulation service reserves, Quality-of-Service (QoS) provisioning for the jobs running at the data centers has not been considered. In this work, we propose an Adaptive policy with QoS Assurance that enables data centers to participate in regulation service programs with assurance on job QoS. Our policy regulates data center power through job scheduling and server power capping. QoS assurance is achieved by applying a queueing-theoretic result to our job scheduling strategy. We evaluate our policy by experiments on a real cluster. Our results demonstrate that the proposed policy reduces electricity costs by 25-56% while providing QoS assurance. On the other hand, the baseline policies cannot meet QoS constraints in 9 of the 14 workload traces tested.

The impact of payer status on hospital admissions: evidence from an academic medical center.

BACKGROUND: There are plenty of studies investigating the disparity of payer status in accessing to care. However, most studies are either disease-specific or cohort-specific. Quantifying the disparity from the level of facility through a large controlled study are rare. This study aims to examine how the payer status affects patient hospitalization from the perspective of a facility. METHODS: We extracted all patients with visiting record in a medical center between 5/1/2009-4/30/2014, and then linked the outpatient and inpatient records three year before target admission time to patients. We conduct a retrospective observational study using a conditional logistic regression methodology. To control the illness of patients with different diseases in training the model, we construct a three-dimension variable with data stratification technology. The model is validated on a dataset distinct from the one used for training. RESULTS: Patients covered by private insurance or uninsured are less likely to be hospitalized than patients insured by government. For uninsured patients, inequity in access to hospitalization is observed. The value of standardized coefficients indicates that government-sponsored insurance has the greatest impact on improving patients' hospitalization. CONCLUSION: Attention is needed on improving the access to care for uninsured patients. Also, basic preventive care services should be enhanced, especially for people insured by government. The findings can serve as a baseline from which to measure the anticipated effect of measures to reduce disparity of payer status in hospitalization.

Detection of dementia on voice recordings using deep learning: a Framingham Heart Study.

BACKGROUND: Identification of reliable, affordable, and easy-to-use strategies for detection of dementia is sorely needed. Digital technologies, such as individual voice recordings, offer an attractive modality to assess cognition but methods that could automatically analyze such data are not readily available. METHODS AND FINDINGS: We used 1264 voice recordings of neuropsychological examinations administered to participants from the Framingham Heart Study (FHS), a community-based longitudinal observational study. The recordings were 73 min in duration, on average, and contained at least two speakers (participant and examiner). Of the total voice recordings, 483 were of participants with normal cognition (NC), 451 recordings were of participants with mild cognitive impairment (MCI), and 330 were of participants with dementia (DE). We developed two deep learning models (a two-level long short-term memory (LSTM) network and a convolutional neural network (CNN)), which used the audio recordings to classify if the recording included a participant with only NC or only DE and to differentiate between recordings corresponding to those that had DE from those who did not have DE (i.e., NDE (NC+MCI)). Based on 5-fold cross-validation, the LSTM model achieved a mean (±std) area under the receiver operating characteristic curve (AUC) of 0.740 ± 0.017, mean balanced accuracy of 0.647 ± 0.027, and mean weighted F1 score of 0.596 ± 0.047 in classifying cases with DE from those with NC. The CNN model achieved a mean AUC of 0.805 ± 0.027, mean balanced accuracy of 0.743 ± 0.015, and mean weighted F1 score of 0.742 ± 0.033 in classifying cases with DE from those with NC. For the task related to the classification of participants with DE from NDE, the LSTM model achieved a mean AUC of 0.734 ± 0.014, mean balanced accuracy of 0.675 ± 0.013, and mean weighted F1 score of 0.671 ± 0.015. The CNN model achieved a mean AUC of 0.746 ± 0.021, mean balanced accuracy of 0.652 ± 0.020, and mean weighted F1 score of 0.635 ± 0.031 in classifying cases with DE from those who were NDE. CONCLUSION: This proof-of-concept study demonstrates that automated deep learning-driven processing of audio recordings of neuropsychological testing performed on individuals recruited within a community cohort setting can facilitate dementia screening.